Inheritance in sample array multitree subdivision
A better compromise between encoding complexity and achievable rate distortion ratio, and/or to achieve a better rate distortion ratio is achieved by using multitree sub-divisioning not only in order to subdivide a continuous area, namely the sample array, into leaf regions, but using the intermediate regions also to share coding parameters among the corresponding collocated leaf blocks. By this measure, coding procedures performed in tiles—leaf regions—locally, may be associated with coding parameters individually without having to, however, explicitly transmit the whole coding parameters for each leaf region separately. Rather, similarities may effectively exploited by using the multitree subdivision.
1. A decoder for reconstructing an array of information samples representing a spatially sampled video information signal, which is subdivided into a multi-tree structure of leaf regions of different sizes by multi-tree subdivision, wherein the multi-tree structure is obtained using a primary sub-division and a sub-ordinate sub-division, such that a tree-root block of the primary sub-division is sub-divided leaf blocks of the primary sub-division which, in turn, form the tree-root blocks of the sub-ordinate sub-division, further sub-divided into leaf regions of the multi-tree structure, each leaf region of the multi-tree structure has associated therewith (a) a hierarchy level out of a sequence of hierarchy levels of the multi-tree subdivision, and (b) coding parameters, which are, for each leaf region, represented by a respective set of syntax elements, the decoder comprising:
an extractor configured to extract from the data stream, using a processor, inheritance information indicating whether inheritance is used, and based on an indication from the inheritance information that inheritance is to be used, at least one inheritance region of the array of information samples which is spatially composed of a set of non-overlapping sub-regions, wherein each of the set of non-overlapping sub-regions has a same resolution as that of the at least one inheritance region,
wherein, responsive to the indication from the inheritance information that inheritance is to be used, the extractor is further configured to extract from the data stream, using the processor, at least one syntax element associated with the at least one inheritance region, the at least one syntax element including an intra-prediction mode syntax element,
wherein the decoder is configured to:
decode, in a residual decoding order, using the processor, a residual signal for each of the set of non-overlapping sub-regions,
copy, using the processor, the at least one syntax element associated with the at least one inheritance region as a corresponding syntax element associated with the respective non-overlapping sub-region,
obtain, in the residual decoding order, using the processor, a separate intra prediction signal for each of the set of non-overlapping sub-regions according to an intra-prediction mode indicated by the intra-prediction mode syntax element, and
reconstruct, using the processor, each of the set of non-overlapping sub-regions by adding the respective intra prediction signal and the respective residual signal.
2. The decoder of claim 1 , wherein the at least one inheritance region corresponds to a first hierarchy level of the sequence of hierarchy levels, the first hierarchy level being lower than a second hierarchy level of the sequence of hierarchy levels with which the set of non-overlapping sub-regions are associated.
3. The decoder of claim 1 , wherein the separate intra prediction signal is obtained using neighboring samples of a reconstructed signal of an already reconstructed sub-region.
4. The decoder of claim 1 , wherein the extractor is configured to extract, from the data stream, at least one syntax element of a predetermined syntax element type representing a coding parameter associated with a leaf region that is not a sub-region of the at least one inheritance region.
5. The decoder of claim 1 , wherein the extractor is configured to, responsive to the indication from the inheritance information that inheritance is not to be used, extract at least one syntax element of a predetermined syntax element type representing a coding parameter associated with the leaf regions of the array of information samples.
6. The decoder of claim 1 , wherein the residual decoding order is a depth-first traversal order.
7. The decoder of claim 1 , wherein the decoder is a hybrid video decoder.
8. The decoder of claim 1 , wherein the decoder is configured to, for each of the set of non-overlapping sub-regions, instead of copying the at least one syntax element, use the at least one syntax element associated with the at least one inheritance region to predict the corresponding syntax element.
9. A method for reconstructing an array of information samples representing a spatially sampled video information signal, which is subdivided into a multi-tree structure of leaf regions of different sizes by multi-tree subdivision, wherein the multi-tree structure is obtained using a primary sub-division and a sub-ordinate sub-division, such that a tree-root block of the primary sub-division is sub-divided leaf blocks of the primary sub-division which, in turn, form the tree-root blocks of the sub-ordinate sub-division, further sub-divided into leaf regions of the multi-tree structure, each leaf region of the multi-tree structure has associated therewith (a) a hierarchy level out of a sequence of hierarchy levels of the multi-tree subdivision, and (b) coding parameters, which are, for each leaf region, represented by a respective set of syntax elements, the method comprising:
extracting from the data stream, inheritance information indicating whether inheritance is used, and based on an indication from the inheritance information that inheritance is to be used, at least one inheritance region of the array of information samples which is spatially composed of a set of non-overlapping sub-regions, wherein each of the set of non-overlapping sub-regions has a same resolution as that of the at least one inheritance region,
responsive to the indication from the inheritance information that inheritance is to be used, extracting from the data stream, at least one syntax element associated with the at least one inheritance region, the at least one syntax element including an intra-prediction mode syntax element,
decoding, in a residual decoding order, a residual signal for each of the set of non-overlapping sub-regions,
copying the at least one syntax element associated with the at least one inheritance region as a corresponding syntax element associated with the respective non-overlapping sub-region,
obtaining, in the residual decoding order, a separate intra prediction signal for each of the set of non-overlapping sub-regions according to an intra-prediction mode indicated by the intra-prediction mode syntax element, and
reconstructing each of the set of non-overlapping sub-regions by adding the respective intra prediction signal and the respective residual signal.
10. The method of claim 9 , wherein the at least one inheritance region corresponds to a first hierarchy level of the sequence of hierarchy levels, the first hierarchy level being lower than a second hierarchy level of the sequence of hierarchy levels with which the set of non-overlapping sub-regions are associated.
11. The method of claim 9 , wherein the separate intra prediction signal is obtained using neighboring samples of a reconstructed signal of an already reconstructed sub-region.
12. The method of claim 9 , further comprising extracting, from the data stream, at least one syntax element of a predetermined syntax element type representing a coding parameter associated with a leaf region that is not a sub-region of the at least one inheritance region.
13. The method of claim 9 , further comprising, responsive to the indication from the inheritance information that inheritance is not to be used, extracting at least one syntax element of a predetermined syntax element type representing a coding parameter associated with the leaf regions of the array of information samples.
14. The method of claim 9 , wherein the residual decoding order is a depth-first traversal order.
15. The method of claim 9 , further comprising, for each of the set of non-overlapping sub-regions, instead of copying the at least one syntax element, using the at least one syntax element associated with the at least one inheritance region to predict the corresponding syntax element.
16. An encoder for encoding, into a data stream, an array of information samples representing a spatially sampled video information signal, which is subdivided into a multi-tree structure of leaf regions of different sizes by multi-tree subdivision, wherein the multi-tree structure is obtained using a primary sub-division and a sub-ordinate sub-division, such that a tree-root block of the primary sub-division is sub-divided leaf blocks of the primary sub-division which, in turn, form the tree-root blocks of the sub-ordinate sub-division, further sub-divided into leaf regions of the multi-tree structure, each leaf region of the multi-tree structure has associated therewith (a) a hierarchy level out of a sequence of hierarchy levels of the multi-tree subdivision, and (b) coding parameters, which are, for each leaf region, represented by a respective set of syntax elements, the encoder comprising:
a data stream inserter configure to:
insert into the data stream, using a processor, inheritance information indicating whether inheritance is used, and based on an indication from the inheritance information that inheritance is to be used, at least one inheritance region of the array of information samples which is spatially composed of a set of non-overlapping sub-regions, wherein each of the set of non-overlapping sub-regions has a same resolution as that of the at least one inheritance region,
responsive to the indication from the inheritance information that inheritance is to be used, insert into the data stream, using the processor, at least one syntax element associated with the at least one inheritance region, the at least one syntax element including an intra-prediction mode syntax element, and
insert into the data stream, using the processor, in a residual coding order, a residual signal for each of the set of non-overlapping sub-regions,
wherein, if inheritance is to be used, the at least one syntax element associated with the at least one inheritance region is to be copied as a corresponding syntax element associated with the respective non-overlapping sub-region,
a separate intra prediction signal for each of the set of non-overlapping sub-regions is to be obtained according to the residual coding order, and an intra-prediction mode indicated by the intra-prediction mode syntax element, and
each of the set of non-overlapping sub-regions is to be reconstructed by adding the respective intra prediction signal and the respective residual signal.
17. The encoder of claim 16 , wherein the at least one inheritance region corresponds to a first hierarchy level of the sequence of hierarchy levels, the first hierarchy level being lower than a second hierarchy level of the sequence of hierarchy levels with which the set of non-overlapping sub-regions are associated.
18. The encoder of claim 16 , wherein the separate intra prediction signal is to be obtained using neighboring samples of a reconstructed signal of an already reconstructed sub-region.
19. The encoder of claim 16 , wherein the data stream inserter is configured to insert, into the data stream, at least one syntax element of a predetermined syntax element type representing a coding parameter associated with a leaf region that is not a sub-region of the at least one inheritance region.
20. The encoder of claim 16 , wherein the data stream inserter is configured to insert, into the data stream, responsive to the indication from the inheritance information that inheritance is not to be used, at least one syntax element of a predetermined syntax element type representing a coding parameter associated with the leaf regions of the array of information samples.
21. The encoder of claim 16 , wherein the residual coding order is a depth-first traversal order.
22. The encoder of claim 16 , wherein the encoder is a hybrid video encoder.
23. The encoder of claim 16 , wherein for each of the set of non-overlapping sub-regions, instead of copying the at least one syntax element, the at least one syntax element associated with the at least one inheritance region is to be used to predict the corresponding syntax element.